Systems Thinking Chapter 3: Cthulu Ate my Game
First a quick note: in the last chapter I talked about applying systems thinking to game systems, but in fact I think it's more productive to apply it to actual play, whether an explicit system underlies it or not. A lot of what happens in gaming is based on those unwritten practices that groups evolve to make games work. I think a goal of good game design is to make good practices explicit, so ultimately we're still talking about game systems when we talk about diagnosing actual play.
The balancing loop is usually experienced as an oscillation around some target that never quite seems to get where you want it to go. As a graph, the balancing loop looks like this:
In the balancing loop, play either tends towards a goal, or oscillates violently around a goal. Damage output in D&D is a good example. There's an upward trend in how much damage a D&D character can dish out each round, but not all characters follow the trend in the same way. For example, a mage may suddenly leap upwards in relative damage capacity when they get a new level of spell, then dip over the next few levels.
You might also get a similar behavior in campaigns where characters need to pay out treasure to get something that helps them get treasure. In extreme cases you have characters alternately too wealthy or completely broke.
Here's a template for the balancing loop:
I'm going to follow standard procedure and put a B inside the balancing loop and an R inside the reinforcing loop to differentiate them.
Performance is some game element that has a desired target.
Gap is the gap between the desired target and actual performance.
Corrective action is how the gap is addressed.
Limiting factor is an optional component that represents the finite limit on performance.
Note that in the reinforcing limit there may be a delay in time between the corrective action and the result, making it harder to figure out what's actually going on.
Actual Play Examples
I ran a Cthulu campaign that was a huge success. I had two groups of two players each. I was getting great play out of my players and having a blast running it. Several other people also wanted to get involved. So I ran a big game with all four players and three visitors. It bombed massively, and my players left. It turned out that they were enjoying all the close person GM attention. When the number of players rose above the threshold where I could attend to each one individually, then nobody was having fun any more. My mistake was the unspoken assumption that more players is better.
Here's this play example as a system diagram:
I can also recall at least one GM who over a course of years had alternately 2-3 groups on the go or was bemoaning his lack of players. I suspect something similar was the cause. Players would often stick around for a while before leaving, making it harder to see the actual cause of dissatisfaction.
I think that good balancing loops can be invaluable for game design. I expect that in almost any case where "balance" is an issue, a balancing loop is involved (or else should be). The conflict system in Universalis might be an example where the loser in a conflict reaps a reward of coins, giving them more narrative control later on.
The balancing loop is usually experienced as an oscillation around some target that never quite seems to get where you want it to go. As a graph, the balancing loop looks like this:
In the balancing loop, play either tends towards a goal, or oscillates violently around a goal. Damage output in D&D is a good example. There's an upward trend in how much damage a D&D character can dish out each round, but not all characters follow the trend in the same way. For example, a mage may suddenly leap upwards in relative damage capacity when they get a new level of spell, then dip over the next few levels.
You might also get a similar behavior in campaigns where characters need to pay out treasure to get something that helps them get treasure. In extreme cases you have characters alternately too wealthy or completely broke.
Here's a template for the balancing loop:
I'm going to follow standard procedure and put a B inside the balancing loop and an R inside the reinforcing loop to differentiate them.
Performance is some game element that has a desired target.
Gap is the gap between the desired target and actual performance.
Corrective action is how the gap is addressed.
Limiting factor is an optional component that represents the finite limit on performance.
Note that in the reinforcing limit there may be a delay in time between the corrective action and the result, making it harder to figure out what's actually going on.
Actual Play Examples
I ran a Cthulu campaign that was a huge success. I had two groups of two players each. I was getting great play out of my players and having a blast running it. Several other people also wanted to get involved. So I ran a big game with all four players and three visitors. It bombed massively, and my players left. It turned out that they were enjoying all the close person GM attention. When the number of players rose above the threshold where I could attend to each one individually, then nobody was having fun any more. My mistake was the unspoken assumption that more players is better.
Here's this play example as a system diagram:
I can also recall at least one GM who over a course of years had alternately 2-3 groups on the go or was bemoaning his lack of players. I suspect something similar was the cause. Players would often stick around for a while before leaving, making it harder to see the actual cause of dissatisfaction.
I think that good balancing loops can be invaluable for game design. I expect that in almost any case where "balance" is an issue, a balancing loop is involved (or else should be). The conflict system in Universalis might be an example where the loser in a conflict reaps a reward of coins, giving them more narrative control later on.
Labels: systems thinking
Tony,
I think I sort of get it, but for clarity, what is the major difference between the reinforcing loop and the balancing loop?
Keep up the good work by the way. I am enjoying the way you make the theory tangible and interesting by making it personal.
Posted by
RogerT |
8:07 PM
In the reinforcing loop, some action pushes performance which increases (or decreases) exponentially.
In the balancing loop, some force pushes performance towards a target. Performance either zooms in on the target, or keeps going under and then over the target alternately.
The key difference is that the balancing loop has a target. It may also have a limiting factor that keeps it from moving exponentially in one direction or the other.
This isn't specifically brought up in the stuff I'm reading, but it looks to me like you also get a balancing loop when you shoot for a reinforcing loop, but keep running up against a limit you haven't diagnosed. i.e. you're trying to get more and more players, but there's a limit that keeps you from getting there.
Posted by
Unknown |
8:38 PM
New reader. This is some facinating stuff though...
I think the key problem in using this stuff is the difficulty of being able to distill the issue down to one or two salient points.
Like: the players were enjoying the DMs attention. Large groups can fail for a whole host of other reasons including, intra-player conflict between players who don’t like each other but are now stuck together, a weakened story theme due to having conflicting character backgrounds, the impact of a new players or group of players play style, and so forth.
Just a thought.
Posted by
Anonymous |
12:52 AM
Yep, you're absolutely right. The more I try to explain this stuff, the more I realize that it's just an approximation. Systems thinking can help clarify what happens when people game, but it's not a quick fix for anything.
Welcome to attacks!
Posted by
Unknown |
8:16 AM
I agree with graf about the difficulty in distilling this down to a few salientpoints, and also send a welcome :. further to add to your comment Tony, I think that systems thinking might be well put to use as a diagnostic aid, to try to solve problems at the table, and perhaps has less utility as a design tool. This is not meant to stifle this discussion in any way. I think that one of the things we don't talk enough about is how to analyze what is going on at the table - and instead often just try another game, which may or may not lead to a solution. It's kind of like treating a disease by trying a different drug instead of analyzing what is going on in the body. I think systems thinking may help us analyze what is going on in the body (in this case the table). I am not sure if it is as much help in what is going on with the underlying rules, although I am willing to admit to being less familiar with it than you and my opinion may very well be changed by future articles.
Posted by
RogerT |
10:05 AM
hmmm... great analogy. I only half agree with you, but I think I need a whole post to explain why...
Posted by
Unknown |
2:38 PM
As an engineer and a retired pen-and-paper gamer, I found your use of systems theory rather interesting. One of the main points of systems theory is that only by looking at the big picture can we find the processes that drive that system. Thus it necessarily ought to be difficult to distill an example down to a few key points.
Any thinking in this manner ought to be directly relevant for game design. For example, how experience points ought to be distributed to obtain an optimal balance of player satisfaction while still keeping their interest.
When I was quite young I played D&D with a great GM. One of his skills was that he could GM one-on-one while playing other player's characters as NPC's. Then, he'd run the same adventure with the other players from their perspective! It made the gaming sessions really interesting, and then when we all met up the adventures would continue where we'd left off. I'm not sure how other GM's cope with varying player participation, but that style certainly worked for us.
Posted by
Shannon |
3:34 PM